Klystrons and other electron beam devices requiring a magnetic field for their operation could be made smaller and lighter if the electromagnet could survive the high temperature bakeout required during the fabrication of the device. There are many high temperature electrical devices, for example, electrical heating elements using ceramic or mica insulation. The requirements for a bakeable electromagnet are different from a heating element, however, in that the conductor in an electromagnet needs to be kept as cool as possible during operation.
The approach to forming an electromagnet on an electron beam tube can be divided into a "wrapped solenoid" approach and a "wound-on magnet" approach. In the "wrapped solenoid" approach, the tube is assembled and baked and then a solenoid is wrapped on the tube using the tube as a spool. In the "wound-on magnet" approach, the electromagnet is a component to be assembled with other components to make a complete tube and then baked.
The "wound-on magnet" design has several serious disadvantages. It is presumed that the tube is tested first in an ordinary solenoid magnet to assure meeting all electrical specifications. Then the device, now representing a substantial monetary investment, is mounted in a winding fixture for application of the magnet turns. The magnet winding operation may or may not be successful. In either case, further testing must be carried out. If the magnet does not yield the desired results, then it must be unwound and a second attempt made. The technique offers no change to check the magnet before it is used. Another disadvantage rests in the fact that cutouts, such as those used for passage of the output waveguide, are not possible. Still further, a special system of cooling might be required to remove coil heat. One system that has been successful with low power linear beam tubes make use of coil circulated in contact with the coils. A separate oil-water heat exchanger is employed.
The "bakeable" magnet calls for the use of certain materials that differ from those used in conventional solenoids. The coil winding insulation must withstand the bakeout temperatures. Metal oxides have been used in some attempts in the past, though the history of such units suggest trouble from turn-to-turn shorts.